Saturday, January 17, 2009

The Globalized Rail Industry

NOTE: We've moved! Visit us at the California High Speed Rail Blog.

Over at the European Tribune DoDo offers an excellent look at how globalization has affected the manufacturing of HSR trainsets. This clearly has implications for our own high speed rail system, since most of the manufacturers who would provide trainsets for California HSR are based in other countries (though I hope and expect they will produce the cars here in CA). Anyhow, it's a very good read and surveys not only the technical aspects of HSR technology but the political considerations behind them, how shifting markets in both Europe and Asia have driven innovation. For example:

Alstom's ambitions in particular were long hampered by SNCF's unwillingness to order the AGV: they preferred the lower price and reliability of the TGV (and the higher capacity of the double-deck Duplex) over (relatively minor) improvements in performance.

The AGV, you'll recall, was the trainset that Fiona Ma rode in 2007, a trip that did much to raise the profile of HSR here in California. Had SNCF been willing to order the AGV it could have made that technology a more readily available option to us in California. True, that option is still there, but it's my belief that the CHSRA will - rightly, in my view - prefer to go with a more commonplace trainset than something untested and new.

Perhaps it will be Spain that pushes the envelope most effectively:

Finally, there is the above mentioned announced threat to everyone from Spain, the Talgo AVRIL. To cut Madrid-Barcelona times to 1h45m(!), the targeted service top speed is 380 km/h. With Talgo's expertise, this should be considered with more seriousness than China's ambition -- but not without doubts. Talgo named its current top product Talgo 350, but it is certified for only 330 km/h (with insufficient power and ride quality among the likely reasons).

380 km/h is about 236 mph, which would achieve the CHSRA's stated goal of 220 mph top speeds on the line. If Talgo can deliver that service to RENFE then it becomes much more likely California can get that as well.

It's likely that all of these will be evaluated when the Central Valley test track is built. But I have to believe that past performance will be a factor as well - if, say, the Talgo AVRIL can reliably accomplish its intended speeds in Spain then it would augment a successful test here in California.

Ultimately this decision has political ramifications, and the state of the global HSR market will play a major role. So go have a look at DoDo's post to get better acquainted with that market.


Anonymous said...

Robert I know you are biased towards Talgo because of your AVE nostalgia, but I fully expect the Kawasaki efSET and the promised AGV Duplex to offer compelling competition. These sets along with the AVRIL (which BTW has an unfortunate connotation) are all in conceptual stages right now, although they may be prototyped and tested by the time the CHSRA opens the bidding process.

Anonymous said...

But of course, if every elite trainset manufacturer had Philippe Mellier's sentiment, they would be wary of exporting their latest technology. That would mean the CHSRA would have to choose only older train sets (e.g. Talgo 350). But it could be seen as a win-win if the CHSRA insists on only well-tested, proven technology.

Robert Cruickshank said...

Honestly, I'm not biased at all. I confess to having hardly any firm understanding at all of train technologies or engineering. My focus has always been on the big picture and the politics of HSR.

I don't have any preferences in terms of train manufacturers, and mentioned the Talgo AVRIL only because it had been mentioned in the source article I linked to. If the CHSRA decides after a period of evaluation that the other trainsets you mentioned are superior, fine by me. I want a train that is reliable and efficient to operate and maintain.

I agree fully that the CHSRA should insist on, as you say, "well-tested, proven technology."

Anonymous said...

I think we will get the ICE3 trainset..We have a Siemens plant right in Sacramento. Thou The Gov and Fiona did ride the French trains. And it might come down to who will help fund the system.

Anonymous said...

BTW love all those beautiful trainsets on the web link..I cant wait to see the first CAHSR set and what livery its in. I do like the concept colors.Well this is alot more fun than Oct when all the deniers were screaming the bond was going down!

Brandon in California said...

When the time comes for trainsets to be vetted, I suspect very much that I will get behind the one that provides reliaable and quick service (doh), along with provide the necessary capacity.

If the ridership projections are reasonable, it seems to me the system will need to be built to provide frequent service AND provide high capacity trainsets. Double-decker cars seem like they will need to be a prerequisite based on my understanding so far.

What degree will infrastructure need to reflect double-decker trainssets... I don't know. However, tunnel design/diameter comes to mind.


If annual demand is anticipated to reach 90-117 million (approximately), daily ridership will range from 250k to 350k(approximately). Maybe a bit more.

Assuming 98 CHSRA train trips are operated per previousl literature, on average 2,500 to 3,300 people will board each train trip. Granted, there will be many ons and offs from end to end and peak load will not reach this number, but, because the main trip pattern will likely be between the major north and south metros, north to south, the number will still be very high.

Because of this, I suspect the design capacity of trainsets should be at least 50% of the 2,500-3,300 range; or 1,250 to 1,650. Maybe more?

Now, if the 98 train trip figure should include additional local/regional commute trips (maybe they're supported by local funding; ie Caltrain-type service), and a better estimate of train trips to be provided is closer to, say 150, then using the same approach as above, then traisnsets should accomodate 800 to 1,100 persons.

That's lower, but still a lot of people.


For comparison purposes, BART's daily weekday ridership is over 350,000. However, they run far more than 98 or 150 trains! It's also a much different style of service, commute vs intercity.

But, at peak times, 10-car trains from Antioch into San Francisco are packed with upwards of 100 people per car, maybe many more (70-ish seats and many standees). Other BART lines and trips are like that too, but I understand the Antioch-SF line is the most heavily used. Upwards of over 1,000 people can be on a BART train at a single moment. They are about 700 feet long.

My understanding of these figures, granted from looking at these from a high elevation wtihout ground-level detail, tell me that design capacity needs to be considered strongly.

Rafael said...

The good news here is that there are multiple vendors to choose from. For maglev technology, there is only one (Transrapid in Germany) with an off-the-shelf product that is already in commercial service (in Shanghai).

I fully expect CHSRA to drive a hard bargain and, that any vendor that wants to make it onto the shortlist will need to commit to setting up an assembly plant in California (or at least the US) if they are selected.

However, let's not forget that the first hurdle in technology selection is getting that all-important "rule of special applicability" from FRA. Let's hope Obama/Biden and others insist on an FRA chief with significant expertise in both passenger and freight rail topics. It is not a choice that should be delegated to Ray LaHood, though it should be his prerogative to name a candidate.

Alon Levy said...

I don't think anyone's worried that the US will copy foreign technology and reexport it. Protectionists are as a rule worried mostly about a) countries that they have to compete with in the present, and b) foreign cultures. (Compare the hysteria in the US about Japanese cars and outsourcing to Mexico and India, with the relative silence about outsourcing to Canada.)

neroden@gmail said...

I am actually sure that any of the standard train manufacturers can provide fine, suitable trains. (Even Bombardier, which hasn't actually made a high-speed trainset yet, has plenty of expertise.)

...*if* the FRA provides the "rule of special applicability" -- or, better, fixes its built-like-a-tank rules, which Caltrain has proven to be worse than European rules even in the case of mixed traffic with poorly signalled freight trains.

That really is the essential matter.

Anonymous said...

Given how important certified 350-360 km/h (220 mph) operation is to the project, speed is a big deal. In terms of speed and availability, the AGV (360 km/h) may be in the best position. All other train sets currently in production have a certified maximum speed of only 330 km/h or less. The next generation 350 km/h+ train sets (AVRIL, efSET, AGV Duplex, HEMU-400x) probably will not be well tested enough for the CHSRA's liking. And I do not expect the CHSRA to "overclock" their trains the way China does with their CRH3 trains.

But let's see whether these technical concerns outweigh the political ones, or vice versa.

Anonymous said...

The only thing I'm concerned so far are just the amenities.

Anonymous said...

Your obsession with Spain is becoming boring.

Anonymous said...

Actually Talgo makes only the passenger cars. The locomotive is made by Bombardier.

Anonymous said...

The AGV is currently in production for use in Italy. By the time Cali comes to make a descision about rollign stock, it will alreayd have notched many miles of service.
The AGV is evolutionarty, not revolutionary in its technology and engirneering. Also, it has the killer advantage of being the fastest train available off-the- shelf. To pick anything else would be to pick a lower service.

Clem said...

@Brandon - annual demand of 90 - 117M sounds pretty lofty, compared to 20M annual airline seats (not all of them filled) offered today between SF - LA regions (all airports). My guess is that single-deckers will do just fine.

Alon Levy said...

High-speed rail is far more convenient than airlines, especially on routes where it can compete on both time and price. In fact, most of the projected ridership will come from reducing auto travel (link).

Besides, we can look at the one pair of cities of comparable size to LA and SF with close economic integration and a good HSR connection, Tokyo-Osaka, which has an annual ridership of about 130 million. Tokyo is twice the size of LA metro and Osaka is twice the size of the Bay Area, but conversely, an LA-SF ticket is expected to cost $55, compared with $135 on a Nozomi train between Tokyo and Osaka, and the CAHSR system will serve more destinations.